¹ Documented worldwide as of 1988 by Gastil, Raymond, Freedom in the World: Political Rights and Civil Liberties, 1988–1989 (New York: Freedom House, 1989)Google Scholar, and traced back to the eighteenth century by Modelski, George, Is America's Decline Inevitable? (Wassenaar: Netherlands Institute for Advanced Study, 1988)Google Scholar. See also Starr, Harvey, “Democratic Dominoes: Diffusion Approaches to the Spread of Democracy',” journal of Conflict Resolution 35 (June 1991)CrossRefGoogle Scholar.

² The literature on this topic is now very substantial. The first empirical reports were Dean Babst, “A Force for Peace,” Industrial Research (April 1972); Wallensteen, Peter, Structure and War: On International Relations, 1820–1968 (Stockholm: Raben & Sjogren, 1973)Google Scholar; and Small, Melvin and Singer, J. David, “The War-Proneness of Democratic Regimes,” Jerusalem Journal of International Relations 1, no. 1 (1976)Google Scholar. Early theoretical contributions, recent empirical reports, and reviews include Rummel, R. J., Understanding Conflict and War, vols. 2, 4, 5 (Los Angeles, Calif.: Sage, 1976, 1979, 1981)Google Scholar; idem, “Libertarianism and International Violence,” Journal of Conflict Resolution 27 (March 1983)Google Scholar; Russett, Bruce and Starr, Harvey, World Politics: The Menu for Choice (New York: W. H. Freeman, 1981), chap. 15Google Scholar; Doyle, Michael, “Kant, Liberal Legacies, and Foreign Affairs,” Parts 1 and 2, Philosophy and Public Affairs, 12, nos. 3—4 (1983)Google Scholar; idem, “Liberalism and World Politics,” American Political Science Review 80, no. 4 (1986)Google Scholar; Maoz, Zeev and Abdolali, Nasrin, “Regime Types and International Conflict,” Journal of Conflict Resolution 33 (March 1989)Google Scholar; Russett, Bruce, Controlling the Sword: The Democratic Governance of National Security (Cambridge: Harvard University Press, 1990), chap. 5CrossRefGoogle Scholar. For a general argument on the necessity for explaining war at the dyadic level, see Most, Benjamin and Starr, Harvey, Inquiry, Logic, and International Politics (Columbia: University of South Carolina Press, 1989)Google Scholar.

³ Spencer Weart, Why Don't Democracies Fight One Another? (forthcoming), suggests that certain types of oligarchies have rarely fought each other in various historical eras, but Zeev Maoz and Bruce Russett and Stuart A. Bremer find no evidence for this in the modern world. See Maoz, and Russett, , “Alliances, Distance, Wealth, and Political Stability: Is the Lack of Conflict among Democracies a Statistical Artifact?” International Interactions 17, no. 3 (1992)CrossRefGoogle Scholar; and Bremer, , “Dangerous Dyads: Conditions Affecting the Likelihood of Interstate War, 1816–1965,” Journal of Conflict Resolution 36 (June 1992)CrossRefGoogle Scholar.

⁴ Huntington, Samuel P., “No Exit: The Errors of Endism,” National Interest 17 (Fall 1989)Google Scholar.

⁵ Weede, Erich, “Extended Deterrence by Superpower Alliance,” Journal of Conflict Resolution 27 (June 1983)CrossRefGoogle Scholar. But Randolph Siverson and Juliann Emmons show that since 1920 democracies have been more likely to ally with one another because they were democratic. See Siverson, and Emmons, , “Birds of a Feather: Democratic Political Systems and Alliance Choices,” Journal of Conflict Resolution 35 (June 1991)CrossRefGoogle Scholar. Mesquita, Bruce Bueno de, The War Trap (New Haven: Yale University Press, 1981)Google Scholar, and Maoz and Russett (fn. 3) show that in the modern world allies are actually more likely to fight each other than to fight nonallies.

⁶ Mueller, John, Retreatfrom Doomsday: The Obsolescence of Major War (New York: Basic Books, 1989)Google Scholar.

⁷ Maoz and Russett (fn. 3); idem, Normative and Structural Causes of Democratic Peace, 1946–1986 (New Haven: Yale University International Security Program, 1992)Google Scholar.

⁸ Most of the authors cited in fn. 2 write predominantly from this perspective.

⁹ Most recently Bremer (fn. 3); Maoz and Russett (fn. 3); and Maoz and Abdolali (fn. 2).

¹⁰ Bruce Bueno de Mesquita and David Lalman present this institutional hypothesis and some confirming evidence; see Mesquita, Bueno de and Lalman, , War and Reason (New Haven: Yale University Press, 1992)Google Scholar, chap. 5. The basic hypothesis goes back at least to Quincy Wright, 4 Study of War (Chicago: University of Chicago Press, 1942), 842–45Google Scholar. Harvey Starr develops the insight that democracy can serve as an indicator of dovishness; see Starr, , “Democracy and War: Choice, Learning, and Security Communities,” Journal of Peace Research 29 (1992)CrossRefGoogle Scholar. David Lake contends that democracies with wide franchises are inherently less imperialistic than are more autocratic states, and while they may fight to resist autarchies, the conjunction of two democracies with low imperialist drive makes them less likely to fight each other. See Lake, , “Powerful Pacifists: Democratic States and War,” American Political Science Review 86, no. 1 (1992)CrossRefGoogle Scholar. T. Clifton Morgan and Valerie Schwebach construct a measure of institutional constraints by which a few democracies are not highly constrained and some nondemocracies are. This allows a preliminary test of institutional versus perceptual constraints if one assumes the latter are found in all democracies and only in democracies. Collinearity between the two variables, however, makes their results inconclusive. See Morgan, and Schwebach, , “Take Two Democracies and Call Me in the Morning: A Prescription for Peace?” International Interactions 17, no. 3 (1992)CrossRefGoogle Scholar.

¹¹ See, e.g., Deutsch, Karl W. et al. , Political Community and the North Atlantic Area (Princeton: Princeton University Press, 1957)Google Scholar; and Destler, I. M. et al. , Managing an Alliance: The Politics of U.S.-Japanese Relations (Washington, D.C.: Brookings Institution, 1976), esp. chap. 5Google Scholar.

¹² Small, Melvin and Singer, J. David, Resort to Arms: International and Civil Wars, 1816–1980 (Los Angeles, Calif.: Sage, 1982)Google Scholar. A good discussion of measurement issues for both democracy and war is Merritt, Richard L. and Zinnes, Dina A., “Democracies and War,” in Inkeles, Alex, ed., On Measuring Democracy: Its Consequences and Concomitants (New Brunswick, N.J.: Transaction, 1991)Google Scholar.

¹³ Segall, Marshall H., “Aggression in Global Perspective: A Research Strategy,” in Goldstein, Arnold P. and Segall, Marshall H., eds., Aggression in Global Perspective (New York: Per-gammon, 1983)Google Scholar.

¹⁴ Percentage calculated from Murdock, George P., “Ethnographic Atlas: A Summary,” Ethnology 6 (1967)CrossRefGoogle Scholar.

¹⁵ For an analysis of the ancient Greek system, see Russett, Bruce and Antholis, William, “Democracies Rarely Fight Each Other? Evidence from the Peloponnesian War” (New Haven: International Security Program, Yale University, 1991)Google Scholar.

¹⁶ Ross, Marc H., “Political Decision-Making and Conflict: Additional Cross-Cultural Codes and Scales,” Ethnology 22 (1983)CrossRefGoogle Scholar.

¹⁷ When it comes to external war, however, we could not make that assumption. The units that fight each other in external warfare are by definition from different cultures, and we cannot assume that the “enemy” has the same degree of political participation.

¹⁸ Murdock, George P. and White, Douglas R., “Standard Cross-Cultural Sample,” Ethnology 8 (1969)CrossRefGoogle Scholar; Ross (fn. 16); Tuden, Arthur and Marshall, Catherine, “Political Organization: Cross-Cultural Codes 4,” in Barry, Herbert III and Schlegel, Alice, eds., Cross-Cultural Samples and Codes (Pittsburgh, Pa.: University of Pittsburgh Press, 1980)Google Scholar, originally published in Ethnology 11 (1972); Murdock (fn. 14). The Murdock/White sample includes one well-described society from most of the two hundred cultural regions of the world. Murdock's atlas (fn. 14) is a much larger list of 862 mostly preindustrial societies. To facilitate the retrieval and coding of information on our sample cases, we used the Human Relations Area Files (HRAF) archive whenever possible.

¹⁹ Ember, Melvin and Ember, Carol R., “The Conditions Favoring Matrilocal versus Pat-rilocal Residence,” American Anthropologist 73 (1971)CrossRefGoogle Scholar; Ember, Carol R. and Ember, Melvin, “Melvin” Resource Unpredictability, Mistrust, and War: A Cross-Cultural Study•, Journal of Conflict Resolution (June 1992)Google Scholar.

²⁰ Meggitt, Mervyn, Blood Is Their Argument: Warfare among the Mae Enga Tribesmen of the New Guinea Highlands (Palo Alto, Calif: Mayfield, 1977), 201Google Scholar.

²¹ The scale ranged from 1 to 5, from less often than once in ten years to “constant,” or occurring at any time of year. We exclude here any cases where the coders' initial ratings for internal warfare frequency were not close. By close we mean that one of the following situations applied: (1) The initial ratings did not disagree by more than one point on the five-point ordinal scale. (2) The initial ratings disagreed by more than one point but did not straddle the boundary between low and high frequency of war, which was predictive of various things in past studies (Ember and Ember ffn. 19, 1971]; Ember, Carol R., “Residential Variation among Hunter-Gatherers,” Behavior Science Research 10 [1975]CrossRefGoogle Scholar; idem, “Men's Fear of Sex with Women: A Cross-Cultural Study,” Sex Roles 4 [1978])Google Scholar, i.e., warfare at least once every two years (high) versus less often (low). (3) One of the first two coders said “don't know” and the third coder's rating was “close” (as denned above) to the other initial coder's rating.

²² Ross, Marc H., “Socioeconomic Complexity, Socialization, and Political Differentiation: A Cross-Cultural Study,” Ethos 9 (1981)CrossRefGoogle Scholar.

²³ Ross, Marc H., “Political Organization and Political Participation: Exit, Voice, and Loyalty in Preindustrial Societies,” Comparative Politics 21, no. 1 (1988)CrossRefGoogle Scholar.

²⁴ Dahl, Robert A., Polyarchy: Participation and Opposition (New Haven: Yale University Press, 1971)Google Scholar.

²⁵ Unpublished data by Melvin Ember show that few of these societies have full-time public officials.

²⁶ In this context we mean something less than formal organizations but probably more than the “stable, valued, recurring patterns of behavior” of Samuel P. Huntington's definition of institutions; see Huntington, , Political Order in Changing Societies (New Haven: Yale University Press, 1968), 12Google Scholar. James March and Johan Olsen say that to be identified as a political actor an institution must have a claim of “coherence and authority”; see March, and Olsen, , “The New Institutionalism: Organizational Factors in Political Life,” American Political Science Review 78, no. 3 (1984), 738Google Scholar. Variable 9 (decision-making bodies) most clearly fits, with 6 (checks) and 8 (consultation) relevant to our concern for regularized checks and balances. Variable 7 (removal of leaders) explicitly invokes institutions at the level of the 2 code, but higher levels of restraint depend, appropriately, on norms and perceptions.

²⁷ Murdock (fn. 14). Coding was initially done independently by two trained assistants. Differences in coding exceeding one scale point were discussed by the two coders, typically in conjunction with the third author (on political participation) and the first author (on warfare). If it was apparent that one coder had missed relevant information or had misinterpreted the coding rules, that coder changed her/his rating; sometimes the coders reached a compromise. Where the original codings were more than one point apart and the coders could not reconcile the discrepancy, the case was omitted from the analysis; where the discrepancy was one point or less, the computations were performed using the midpoint between the two codings. The assistants did not know our hypotheses; the third author had no knowledge of the frequency of warfare in the societies.

²⁸ Napoleon Alphonseau Chagnon, “Yanomamo Warfare, Social Organization and Marriage Alliances” (Ph.D. diss., University of Michigan, Ann Arbor, 1966); and idem, Yano-mamo: The Fierce People (New York: Holt Rinehart, 1968)Google Scholar.

²⁹ Hirschman, Albert O., Exit, Voice, and Loyalty (Cambridge: Harvard University Press, 1970)Google Scholar. In a personal communication Spencer Weart has indicated that fission/exile is a clear indication of the breakdown of dissent in ostensibly democratic societies that sometimes do fight each other.

³⁰ Images of a divinity are seen through a glass most darkly, if at all. Hence those images would seem to derive from immediate relationships. Where socialization in the first years of life is strict and punitive, such behavior is typically imputed to the supernatural. Similarly, images of political jurisdiction in the supernatural world tend to parallel those in political jurisdictions above the family. See Lambert, William, Triandis, Leigh, and Wolf, Margery, “Some Correlates of Beliefs in the Malevolence and Benevolence of Supernatural Beings: A Cross-Cultural Study,” Journal of Abnormal and Social Psychology 58 (1959)CrossRefGoogle Scholar; and Swanson, Guy E., The Birth of the Gods (Ann Arbor: University of Michigan Press, 1969)Google Scholar. In international relations, Alexander Wendt argues that beliefs about whether other states will behave according to the principles of realist anarchy are socially constructed, but he attends to learning from behavior in the interstate system rather than from domestic politics. See Wendt, , “Anarchy Is What States Make of It,” International Organization 46 (1992)CrossRefGoogle Scholar. For evidence that evaluations of domestic politics strongly influence orientations toward behavior internationally, see Lumsdaine, David, Ideals and Interests: The Foreign Aid Regime, 1949—1986 (Princeton: Princeton University Press, 1992)Google Scholar.

³¹ Tuden and Marshall (fn. 18).

³² See Carol Ember for earlier work linking type of war to population size; Ember, , “An Evaluation of Alternative Theories of Matrilocal versus Patrilocal Residence,” Behavior Science Research 9 (1974)CrossRefGoogle Scholar.

³³ Ibid. On externalization in modern states, see Levy, Jack, “The Diversionary Theory of War: A Critique,” in Midlarsky, Manus, ed., Handbook of War Studies (Boston: Unwin Hy-man, 1989)Google Scholar.

³⁴ The complete data set will be published separately by the authors in Behavior Science Research.

³⁵ Labovitz, Sanford, “The Assignment of Numbers to Rank Order Categories,” American Sociological Review 35 (1970)CrossRefGoogle Scholar.

³⁶ Another analysis also included a control for patrilocal residence (the pattern whereby a couple lives with or near the husband's parents). Ember and Ember (fn. 19, 1971) contend that residence is a consequence of war, but others consider it a cause because it is an indicator of fraternal interest groups, which are presumed to increase the likelihood of internal war. See, e.g., Otterbein, Keith and Otterbein, Charlotte, “An Eye for an Eye, a Tooth for a Tooth: A Cross-Cultural Study of Feuding,” American Anthropologist 67 (1965)CrossRefGoogle Scholar; and Divale, William, “Migration, External Warfare, and Matrilocal Residence,” Behavior Science Research 9 (1974)CrossRefGoogle Scholar. When included with the above variables, patrilocal residence proved far from significant, partly because, as Carol Ember (fn. 32) has shown, matrilocal societies tend to be significantly smaller than patrilocal ones, and so residence and population are coUinear. The analysis also suggests that residence is coUinear with a few of the participation variables.

³⁷ Standard procedure is to show both equations, with and without outliers. See Lewis-Beck, Michael, Applied Regression: An Introduction (Newbury Park, Calif.: Sage, 1980), 40CrossRefGoogle Scholar.

³⁸ Ross (fn. 22).

³⁹ When either variable 6 or variable 11 was dropped, variable 8 retained its previous sign and approximate significance level; the same occurred for variable 11 when either 6 or 8 was dropped. Variable 6 was more unstable: weakly significant with a positive sign in the absence of 8; weakly significant with a negative sign in the absence of 11.

⁴⁰ In the contemporary international system, see Maoz and Russett (fn. 3), who find that stable democratic systems are more likely to be peaceable toward other democracies than are unstable ones.

⁴¹ Chagnon (fn. 28, 1966), 44.

⁴² Tilly, Charles, Coercion, Capital, and European States, AD 990–1990 (Cambridge, Mass.: Basil Blackwell, 1990)Google Scholar.

⁴³ Data for the Murdock/White sample come from the project reported in Ember and Ember (fn. 19, 1992). The assistants in the project described here coded external warfare for the cases from Murdock (fn. 14).

⁴⁴ Among modern nation-states the evidence that internal violent conflict and external conflict are systematically related is at best ambiguous. See Michael Stohl, “The Nexus of Civil and International Conflict,” and Zinnes, Dinna, “Why War? Evidence on the Outbreak of International Conflict,” both in Gurr, Ted Robert, ed., Handbook^ of Political Conflict (New York: Free Press, 1980)Google Scholar.

⁴⁵ Spoehr, Alexander, Majuro: A Village in the Marshall Islands (Chicago: Natural History Museum, Fieldiana, Anthropology, vol. 39, 1949)Google Scholar; Arno Senftt, Die Marshall-Insulaner (The Marshall Islanders), in Steinmetz, S. R., ed., Rechtsverhaeltnisse von einegeborenen Volkern Africa und Ozeanien (Berlin: Verlag von Julius Springer, 1903)CrossRefGoogle Scholar, translated in 1942 for the Yale Cross-Cultural Survey; Erdland, P. August, Die Marshall-Insulaner, Leben und Siotte, Sinn und Religion eines Suedsee-Volkes (The Marshall Islanders: Life and Customs, Thought and Re ligion of a South Seas People) (Milnster: Anthropos Bibliothek Ethnological Monographs, 1914)Google Scholar, translated in 1942 and 1961 for the Yale Cross-Cultural Survey and HRAF; Wedgewood, Camilla, “Notes on the Marshall Islands,” Oceania 13, no. 1 (1942)Google Scholar.

⁴⁶ Spoehr (fn. 45), 32.

⁴⁷ Nordenskiold, Erland, An Historical and Ethnological Survey of the Cuna Indians, ed. Wassen, Henry (Goteborg: Goteborgs Museum, Etnografiska Avdelningen, 1938)Google Scholar; Fred Mc-Kim, San Bias: An Account of the Cuna Indians of Panama. The Forbidden Land: Reconnaissance of Upper Bayano River, Republic of Panama, in 1936, ed. Henry Wassen (Goteborg: Etnografiska Museet, 1947); Stout, David B., San Bias Cuna Acculturation: An Introduction (New York: Viking Fund, 1947)Google Scholar.

¹⁸ Ross (fn. 16), 176–81.

⁴⁹ Ibid.

⁵⁰ Tuden and Marshall (fn. 18), 120–21.